Non-random utilization of Ig gene segments has been observed frequently at the level of the expressed peripheral repertoire, but the reasons for the biased peripheral V gene utilization are not known. Although much of the peripherally observed over-expression is likely to be due to antigenic selection, we also propose that the V(D) J rearrangement process itself imposes many biases upon the initial repertoire. In this proposal we will dissect the normal human light chanin repertoire. The effects antigenic selection and rearangement biases will be distinguished. The realtive frequency with which individual Vk genes rearrange will be determined, and then experiments will be performed to uncover the reasons why particular gense are not equally used. Factos which might contribute to non-equal rearrangement frequencies, including chromosomal locaiton, proximity to the J region, inversional vs. deletional rearrangement, variations from the consensus RSS, promoter elements, splice sites, and relative transcriptional competency, will be analyzed. Junctional diverisity will also be analyzed for kappa and lambda junctions. The human kappa locus is the optimal locus to study such questions since(1) every Vk gene and its flanking DNA has been sequenced, (2) all Vk genes are mapped, and (3) the most of the Vk locus was duplicated and inverted, somost genes have a very similar copy which is presen in inverted orientation at a much further distance form the J region. Deficiencies in the Ig repertoire, due to allelic or missing genes, or to minor sequence variations in critical regions of coding or regulatory regions, can have a significant impact on specific immune responses and therefore on disease susceptibility. We have identified a new defective allele of A2 in Navajos. We will address the hypothesis that this new allele may play a rol;e in the increased susceptibility of Navajos to H. influenzae type b (Hib) disease. The new allele has an altered RSS, and we directly tesst the affect of this change using a recombination substrate. THe general information obtained i the previous experiements will be applied to the analysis of this specific model immune response. We will test the hypothesis that the ani-Hib response is slow to arid\se in ontogeny because the A2 gene rearranges infrequently due toit distal locaiton and because the invariant V-J junciton seen in all anti-Hib antibodies is rarely made. We will also further analyze the diversity in the ani-hib response through the use of combinatorial libraries, correlating affinity with the Vh, Vk and CDR3's which are used.